MineralsEngineering,Vol.9, No. I, pp.

15-22, 1996
Copyright© 1995ElsevierScienceLtd
Pergamon Printed in GreatBritain.All rightsreserved
0892--6875(95)00128-X 0892-6875/96 $15.00+0.00

RECOVERY OF COPPER FROM LEACHING RESIDUAL SOLUTIONS

BY MEANS OF A HOLLOW-FIBER MEMBRANE EXTRACTOR

F. R. VALENZUELA, C. BASUALTO, C. TAPIA, J. SAPAG and C. PARATORI

Laboratorio de Operaciones Unitarias e Hidrometalurgia,

Facultad de Ciencias Quimicas y Farmac6uticas,
Universidad de Chile, Santiago, P.O.Box 233, Chile
(Received 26 July 1995; accepted 28 September 1995)

ABSTRACT

A process of recovery of copper from acid leach residual solution with 5-

dodecylsalicylaldoxime dissolved in n-hexane was carried out at 30°C in a countercurrent
tubular membrane extractor using a hollow fiber as solid support. The effect of the
extractant concentration and the pH of the feed aqueous solution on the metal transfer
rate and extent of extraction were investigated. It was found that both the flux of metal
and extraction extent were highly influenced by the extractant concentration, and slightly
by the p H of feed solution. No co-transport of other metallic ions was observed over a
p H value of 1.9, allowing attainment of a copper-loaded organic solvent free of
impurities.

Keywords
HydrometaUurgy; Ion exchange; Solvent extraction

INTRODUCTION

The feasibility of separation and enrichment of metals with liquid membranes is one of the most important
technological advances in the area of hydrometallurgy [1,2]. Two kinds of liquid membranes are well
known: liquid surfactant membranes which consist of a water-in-oil-in-water emulsion, and supported liquid
membranes, consisting of a thin highly porous support, whose pores are filled with the organic solvent
which contains an extractant as selective cartier for the metal to be recovered [3,4].

Practical applications of liquid membranes include diverse fields such as gas separation, nuclear waste
processing, desalination of sea water, and others [5,6]. The application of a membrane extractor made of
a hollow fiber as a microporous solid support has been particularly successfully utilized for recovery of
metals from aqueous solution, such that is possible to predict that this technology will be competitive with
other recovery processes, such as ion exchange and conventional solvent extraction using mixer-settler
extractors [7,8].

It has been well documented that the employment of this kind of reactor offers several advantages compared
with the current solvent extraction process in mixer-settlers, particulary huge savings in inventory of
expensive solvent, smaller reactor size and shorter processing times. No direct mixing between aqueous and
organic phases occurs, avoiding crud formation and phase entrainment difficulties.

Several studies on selective separation and concentration of metallic ions with hollow fiber-type supported
liquid membranes have recently been reported [9-12]. Results and analysis of extraction kinetics of copper

9:s-a 15
16 F.R. Valenzuelaet al.

from pure laboratory solutions with a sulfonarnidoquinoline extractant in a hollow fiber membrane extractor
have been reported [ 13]. The rate of copper extraction was analysed by a heterogeneous interfacial reaction
model which considers interfacial adsorption of extractant at the interface between both phases [ 14]. The
present studies, however, are concerned with the kinetics and mechanism of metal extraction from synthetic
solutions prepared in the laboratory.

In this work, the recovery of copper from a dilute solution produced in a mining process is studied using
a hollow fiber membrane extractor. The solution is an acid leach residual solution of a Chilean molybdenite
concentrate, which contains variable but significant concentrations of copper, as well as molybdenum,
rhenium, some iron and arsenic, and other non valuable metals [15,16]. The aim of the present study is to
selectively extract most of the copper present in this solution by means of hollow fiber membrane extractor.
This research will form the basis of the development of a pilot-scale membrane extractor to be designed
in the near future.

EXPERIMENTAL

Materials

LIX-860 (5-dodecylsalicylaldoxime) from Chile Harting-Henkel was used without further purification as the
organic extractant of the liquid membrane, in a wide range of concentration, n-Hexane of commercial GR
grade was used as the diluent in all experiments. No phase modifier reagent was utilized during the
extraction tests. Feed solutions consisted of residual solution from nitric acid leaching processing of a
Chilean molybdenite concentrate. The initial concentration of metals in the solution used in this study were
0.57 g/L Cu(II), 4.0 g/L Mo(VI), 0.30 gFL Fe, 0.30 g/L Re(VII) and minor quantities of other metals. It is
a greenish solution, whose pH averages 0.5-1.0 according to prior treatment, with a density of 1.05 g/mL
at 20°C.

The hollow fiber membrane module was formed by microporous polytetrafluoroethylene fibers (Japan Gore
Tex Inc.) inserted in a glass shell. Some characteristics of the fibers are shown in Table l.

TABLE 1 Characteristics of Hollow Fiber

Porosity 60 %
Maximum Pore Size 2.0 larn
Inner Diameter 0.8 mm
Outer Diameter 1.8 mm
Wall Thickness 0.5 mm
Length 270 m m

Procedure

The aqueous residual leach solution was fed through the inner side of the fiber and the organic solution
through the outer side counter-currently, using peristaltic Masterflex pumps. Conditioning of the feed
solution was carried out by filtering, in order to separate coarse solid particles which could obstruct the pore
structure of membrane.

Initially, several experiments were performed, allowing the feed and the organic solution to flow rapidly
through the porous wall of the fiber in a continuous mode in order to measure the flux of metal through the
membrane. Later experiments were carried out with both the feed and the organic phase at a constant flow
rate, in a recycling mode, in order to check the viability of the selective extraction of Cu(II), continuously
adjusting the pH of feed solution to an appropriate value for metal recovery.
 Recovery of copper from leaching residual solutions 17

The experimental setup was described in a previous paper [13]. Samples of both diffusates were taken at
intervals, to measure the concentration of metals by atomic absorption in a GBC-902 spectrophotometer.

RESULTS AND DISCUSSION

Measurements of flux of metal

The average transfer rate of copper, JM was obtained from a mass balance in the aqueous phase:

rcdi 2 d C M
-u = n di J M (1)
4 dL

u being the lineal velocity of the aqueous phase, d i the inner diameter of the fiber and dCM/dL the variation
of metal concentration along the length (L) of the fibers.

Therefore the flux of metal can be expressed as follows:

u4 (2)

where JM is expressed in mol/cm2.sec, and CM in mol/dm 3. Suffixes "in" and "out" denote inlet and outlet
of extractor respectively. LIX 860 contains in its structure an acid hydrogen-ion which is exchanged with
copper according to the following extraction chemical reaction:

Cu 2 . ~aq)+ 2 nX(org ) = CuX2(org ) + 2 H ÷ (aq) (3)

HX being the organic chelating extractant molecule.

The effect of the LIX-860 extractant concentration and initial pH of feed solution on the metal transfer rate
is shown in Figure 1. In the pH range studied, it was the increase of carrier concentration in the liquid
membrane that caused the greater transfer rate of metal. The observed flux of metal was found to be slightly
dependent on the pH of the feed solution. LIX-860 actually is one of the strongest copper extractants
capable of complexing Cu(II) even around the 0.5 pH value. These copper flux values are somewhat higher
than those obtained by using a mixer-settler solvent extraction system, considering the minimum inventory
of organic solvent utilized to circulate through the membrane wall, and fairly similar to those obtained in
similar experiments using laboratory synthetic solutions.

In previous work [17], we analysed the kinetics and extraction mechanism of copper with salicylaldoxime
- - the active species of LIX-860 - - using the heterogeneous interfacial reaction model to calculate the
adsorption equilibrium constants and interfacial extraction rate constant. The equations of this model, which
has been explained in detail previously [13,14] are:

HXorg ~ HXad , KH x (4)

18 F.R. Valenzuela e t al.

+
Cu 2÷o~1 + HXad ~ CuR oa + H+ ~
K1 (5)

CuX ÷o~ + HXaa ~ CUX2org ÷ a + ,K 2- (6)

kr
Assuming that Eq.(6) is the rate-determining step, the interfacial reaction rate, which under steady-state
conditions must be equal to the mass transfer rate of metal species, can be expressed as follows:

JM = kjC cuiC 2HxiK 2HxK1 (7)

( 1 +KHxSHxCHxi +KHxSHxK1CCuiCHXiC-1H)2

The values of constants calculated by the model for the oximic extractant are shown in Table 2. These
constants are being used to correlate the experimental and the calculated results by use of this interfacial
reaction model. A complete development of this analysis will be the subject of a further report.

-8

-9

A A
Ill W W v
O
¢1=0 LIX-§60

-10
O 1.0 M
0.IM
[] 0.01 M
0.001 M

-11 i I i i
0.5 1.0 1.5 2.0 2.5 3.0
pI'I
Fig.1 Dependence of copper transfer rate on pH of feed solution
and concentration of extractant LIX-860.
Copper content at feed solution: 0.57 g/L

T A B L E 2 E q u i l i b r i u m a n d Rate Constants

KHX' adsorption equilibrium constant = 6.67.10 -7 [cm]

SHX, interfacial area occupied by mole = 6.30.108 [cm2/mol]

K], equilibrium constant in equation 5 = 0.664 [--]

kf, forward reaction rate in equation 6 = 5.93.101° [cm2/mol.s]

 Recoveryof copperfrom leachingresidual solutions 19

Selective recovery of copper

The study of selective recovery of copper from the residual leach solution was carded out by performing
different experiments in a continuous recycling mode.

The influence of two variables were investigated for studying the selectivity of extractant on the recovery
of copper compared with that of molybdenum, rhenium and iron: the pH of aqueous solution and the
concentration of extractant in the organic liquid membrane. The results are shown in Figure 2.

From Figure 2 the following information is observed: the extent of extraction of copper is highly dependent
on a higher molarity of carrier extractant in the liquid membrane. A moderate increase in copper transport
was observed by decreasing the acidity of feed solution.

It is observed from Figure 2 that the concentration of Cu(II) in the feed solution can be lowered significantly
as it is increased in the loaded organic solvent, thus showing the viability of the transport of metal by the
supported liquid membrane.

100 0 1M.pdl.77,1 '

• 1M,pH1.38
. V 0.1M,pH2.2
~'~ 80 ~" [] O.IM.pH1.2
-~ • .

0
0 50 100 150 gO0
Time (rain)
Fig.2 The effect of extractant molarity and pH of feed solution on copper extraction extent
Initial aqueous phase: Cu(II) = 0.57 g/L

Experiments were also performed in order to measure the transport of other metals in the residual leach
solution through liquid membrane, with the organic solvent prepared with LIX-860 in n-hexane. The
permeability to copper was very much greater than the transport observed for all other metallic ions. The
extent of extraction of each metal with the oximic carrier was calculated, and are shown in Table 3, which
provides evidence for the feasibility of separation of copper from the main impurities of the leach solution,
including molybdenum which is present at a high level of concentration, these metals not being transported
through the membrane by the carrier in the experimental conditions tested in this study. Molybdenum (VI)
is scarcely co-transported when the pH of the solution is over 1.9, because in such a pH range no cation
species of this metal exists in aqueous solution, the only state capable of being extracted by LIX reagents
[15,161.

No rhenium load was detected, possibly as in acid aqueous solution this metal exists only as perrhennate
anionic species, not extractable by the oximic extractant [18]. Little co-transport of iron was detected in the
examined range of pH, and no arsenic coextraction was measured.
20 F.R. Valenzuelaet al.

TABLE 3 Metals Extraction Extents as a function of initial pH for 1.0 M LIX-860

Aqueous phase initial concentrations (g/L):
Cu=0.57, Mo=4.0, Re=0.30, Fe=0.30, As=0.40

Extraction Extent (%)

pH Cu Mo Re Fe As

0.77 63.1 26.2 n.d. n.d. n.d.

1.38 73.2 12.4 n.d. n.d. n.d.

1.90 81.4 3.10 n.d. n.d. n.d.

2.51 85.3 n.d. n.d. n.d. n.d.

n.d. : not detected

These results indicate that copper can be selectively transported from acid leach residual solution through
a supported liquid membrane using LIX-860 as a suitable carrier. In fact, over 80% of the feed solution
copper was transfered to the effluent loaded organic solution, ready to be driven to the stripping module
extractor free of impurities.

The transport process proceeds as follows: the protonated carrier diffuses through the liquid membrane to
the feed solution. At the phase boundary, the inner wall of the fiber, the carrier is deprotonated, a copper
ion being added at the same time.

The carrier, which is now loaded with copper ions, then diffuses back through the bulk liquid membrane,
which should be delivered to the stripping module where the carrier is regenerated, ready for the copper ions
collected in the strip liquor to go back to the extraction module. Practically, the liquid membrane acts as
a chemical pump by transporting arid concentrating metal ions from lower- concentration solutions.

It was demonstrated that the polymer film like polytetrafluoro-ethylene used in these experiments,
characterized by high porosity and great mechanical and chemical stability, can be used as a suitable support
material. The thinner the film the greater the flux of metal producing adequate yields.

However, the employment of an appropriate carrier should be considered in the process design. It must be
highly hydrophobic which improves the wetting of the porous polymer support, thus achieving good stability
of the liquid membrane, and form a low viscosity organic solution to ensure a high diffusion rate in the
membrane.

For practical applications, the way in which membranes are designed strongly influences both the operation
and the economics of the process. Undoubtely, a potentially more attractive configuration should consider
several hollow fiber membranes in a tube-and-shell extractor. In operation, the organic complexing carrier
agent must be incorporated into the walls of the fiber, the strip solution flowing on the outside of the fiber,
and the feed solution flowing down inside the fibers. In the pilot-scale extractor design being utilized in the
next stage of this work, a module, or set of modules, will be arranged in this way to try to remove most
of the copper from a continuous feed stream such as residual leach solution.

However, the results of the present study show that the liquid membrane process is a promising alternative
to current solvent extraction and ion-exchange processes, especially for metal recovery from dilute solutions,
being simpler than the conventional processes as the extraction and stripping operation are combined and
considerably less solvent is required.
 Recoveryof copperfrom leachingresidual solutions 21

CONCLUSIONS

A process of recovery of copper from acid leach residual solution has been studied using a hollow fiber
membrane extractor. Measurements of extraction extent and flux of Cu(II) with 5-dodecylsalicylaldoxime
were carried out at 30°C. It was found that the average transfer rate of copper varied proportionally to the
extractant concentration in the liquid membrane and moderately with the pH of the aqueous feed solution.
The permeability of the liquid membrane to copper was very much higher than the transport observed for
all contaminant metallic ions. Molybdenum was co-transported below a pH value of 1.90. No coextraction
of rhenium, iron and arsenic was observed.

ACKNOWLEDGEMENTS

The authors wish to express their thanks to The National Fund of Science and Technology (FONDECYT
Project N°1940456), for the financial assistance given to this work. The authors also thanks Chile Henkel-
Harting Chem. Co. for samples of LIX-860 and to Japan Gore Tex Inc. for supplying hollow fibers.

REFERENCES

.
Babcock, W.C., Baker, R.W., Kelly, D.J. & La Chapelle, E.D., Coupled transport membranes for
uranium recovery, Proc. ISEC'80, 80 (1980).
2. Alexander, P.R. & Callahan, R.W., Liquid-Liquid extraction and stripping of gold with
microporous hollow fiber, J. Membrane Sci., 35, 57 (1987).
3. Li, N.N. & Shrier, A.L., Recent development in separation science, Vol.1, CRC Press, 163-174
(1972).
4. Melling, J., Liquid membrane processes in hydrometallurgy: a review, Warren Spring Lab., L.R.,
1-18 (1979).
5. Lonsdale, H.K., The growth of membrane technology, J.Membrane Sci., 10, 81 (1982).
6. Cooney, D.O. & Jin, C.L., Solvent extraction of phenol from aqueous solution in a hollow fiber
device, Chem. Eng. Commun., 37, 173 (1985).
7. Kim, B.M., Membrane-based solvent extraction for selective removal and recovery of metals, J.
Membrane Sci., 21, 5 (1984).
8. Matsumoto, M., Tsutsumi, Y., Kondo, K. & Nakashio, F., Copper extraction with chelating reagent
in a hollow-fiber membrane extractor. J.Chem. Eng. Japan, 23, 233 (1990).
9. Sato, Y., Kondo, K. & Nakashio, F., Extraction kinetics of molybdenum with 2-ethylhexyl
phosphonic acid mono-2- ethylhexyl ester in membrane extractor using a hollow fiber, J. Chem.
Eng. Japan, 22, 200 (1989).
10. Alonso, A., Ortiz, M., Galan, B. & Irabien, A., Viability and stability study of the recovery of
Cr(VI) with hollow fiber supported liquid membranes, Lat. American Applied Research, 23, 179
(1993).
11. Goto, M., Kubota, F., Miyata, T. & Nakashio, F., Separation of yttrium in a hollow fiber
membrane, J.Membrane Sci., 74, 215 (1992).
12. Valenzuela, F. & Basualto, C., Extracci6n por solventes de cobre mediante membranas lfquidas,
Minerales, 48, 23 (1993).
13. Valenzuela, F., Nakashio, F. & Kondo, K., Extraction kinetics of copper with 8-OSAQ in a
membrane extractor using a hollow fiber, Lat. American Applied Research, 24, 33 (1994).
14. Yoshizuka, K., Kondo, K. & Nakashio, F., Kinetics of copper extraction with N-8-Quinolyl-p-
dodecylbenzenesulfonamide, J.Chem. Eng. Japan, 18, 163 (1985).
15. Valenzuela, F., Poblete, S., Sapag, J., Tapia, C. & Basualto, C., Recovering molybdenum (VI) by
an alkyl phosphonic extractant from acid high copper-containing solutions, Bol. Soc. Chil. Quire.,
40, 25 (1995).
22 F.R. Valenzuelaet al.

16. Valenzuela, F., Andrade, J., Sapag, J., Tapia, C. & Basualto C., The solvent extraction separation
of molybdenum and copper from acid leach residual solution of chilean molybdenite concentrate,
Minerals Engineering, 8, N°8, 893-904 (1995).
17. Valenzuela, F., Basualto, C., Sapag, J. & Tapia, C., Extracci6n de Cu(II) con una salicilaldoxima
en un reactor continuo basado en membranas liquidas de soporte s61ido., Proceedings XXI
Latinoamerican Congress of Chemistry, 1, 247 (1994).
18. Sato, T. & Sato, K., Liquid-liquid extraction of rhenium (VII) from hydrochloric acid solutions
by neutral organophosphorus compounds and high molecular weight amines, Hydrometallurgy, 25,
281 (1990).